Boolean indexing filters a pandas DataFrame by matching a Boolean Series against its ordered rows. Pandas keeps rows whose corresponding value is True and discards the rest. section 6.3.1
A dashboard filter applies the technique in a few steps:
- Select a column with df[col]. section 6.3.1
- Use isin(values) to produce one True or False value for each row. section 6.3.1
- Use df[df[col].isin(values)] to retain only rows matching the user’s selections. section 6.3.1
- Skip the filter when values is empty, preserving the requirement that unselected fields impose no restriction. section 6.3.1
Filtering by a date range
Convert the boundary dates to pandas datetime values, then combine two Boolean conditions with pandas’ element-wise AND operator: df[(df['Day'] >= dt_start) & (df['Day'] <= dt_end)]. This keeps rows between the selected start and end dates, inclusive. section 6.3.2
def apply_filters(df, filters):
for col, values in filters.items():
if values:
df = df[df[col].isin(values)]
return df
Inspect the compact apply_filters implementation that applies boolean indexing across multiple selected columns. section 6.3.1
